Quote from: Taek on September 28, 2015, 10:04:30 PM

Quote from: super3 on September 27, 2015, 05:21:57 PM

Quote from: Taek on September 27, 2015, 04:29:16 PM

Quote from: akaman on September 26, 2015, 10:13:11 PM

Quote from: firedrop on September 25, 2015, 10:23:28 AM

Thanks Taek. Sia seems an interesting and very promising product.
Will be waiting for the release of version 1.0

There is no reason to wait for version 1.0.

Sia is already an interesting and useful protocol. You can already upload files to the Sia network, with the knowledge your files are encrypted and stored redundantly across multiple hosts. The price for this service is currently much cheaper that most, if not all, other storage providers.

In fact, I'd say now is the best time to get in. The Siacoin (SC) used to pay for storage on the Sia network is currently also very, very cheap.

The protocol and client are both still fairly unstable. Great if you are comfortable with bugs, and with the chance of losing money and files because of early-stage instability.

Things get better every iteration, but there is still a lot that could be improved. But it's definitely worth trying out with trivial money and with files you've got backed up in places other than Sia.

You guys are using Reid-Solomon right? What are your parameters?

Right now we're using 2 data pieces with 10 parity pieces. That's a redundancy of 6. By November we're looking at switching to 3 data pieces with 12 parity pieces, for a redundancy of 5. We'll keep raising the numbers and reducing redundancy as the host churn drops.

Ok, that makes sense. Running simulations on our Test Group B data. Will plug that in.